An AWS issue is a problem with Amazon Web Services that stops you from using the cloud resources you need

When you work with AWS — whether you are running a website, storing data, or building an process — you depend on AWS servers and systems to stay running. An AWS issue is any problem that interrupts that service. It might be that a specific AWS region (a geographic location where AWS runs servers) goes down. It might be that a particular service, like the database tool RDS or the storage service S3, stops responding. It might be that your own configuration is broken, or that AWS itself has a bug affecting how your setup works.

The word "issue" covers everything from a five-minute hiccup that you barely notice to a multi-hour outage that takes your entire process offline. Understanding what kind of issue you are facing — and where the problem actually lives — is the first step to fixing it or working around it.

Key Takeaways

  • AWS issues can originate from AWS infrastructure itself, from your own misconfiguration, or from problems in the code or services you have built on top of AWS.
  • AWS publishes real-time status information on the AWS Health Dashboard, which shows which services and regions are affected right now.
  • Most AWS issues are regional, meaning they affect only one geographic location — your process may still work if you have resources in another region.
  • Your AWS bill and your ability to use services may be affected differently depending on whether the issue is on AWS's side or yours.

Where AWS issues actually come from

Not every problem that happens while you are using AWS is an AWS issue. The problem might be on your side. You might have misconfigured a security group (a set of rules controlling which traffic can reach your servers), or you might have run out of the resources AWS allows you to use in your account. You might have deleted something by accident, or your code might have a bug that causes it to crash.

A real AWS issue is one where AWS's own systems are not working as they should. This might be a hardware failure in one of AWS's data centers, a software bug in an AWS service, a network problem between AWS regions, or a problem with the underlying infrastructure that AWS depends on. AWS also performs maintenance — updates and repairs to keep systems running smoothly — and during that time, services may be unavailable or degraded.

The hardest issues to diagnose are the ones that sit in the middle: your code is fine, your configuration looks right, but something about how AWS is behaving has changed. This might be a quota you did not know you had, or a service limit that you have hit, or a change in how AWS handles a particular scenario.

How to tell if the problem is on AWS's side

The fastest way to check is the AWS Health Dashboard, which you can reach from your AWS console. It shows you in real time which AWS services are having problems and which regions are affected. If the dashboard shows a green light for the service you are using, the issue is probably on your side. If it shows yellow (degraded performance) or red (service down), then AWS is having a problem.

You can also check the AWS Status Page at status.aws.amazon.com, which is public and does not require you to log in. This page shows the same information and also displays a history of recent issues and how long they lasted.

If the dashboard is green but your service is still not working, start by checking your own configuration. Look at your security groups, your IAM permissions (which control who can do what in your account), and your service quotas. You can view your current quotas in the Service Quotas console. If you have hit a quota, you can request an increase, though that usually takes a few hours or days.

Regional issues versus account-wide issues

AWS operates in multiple regions around the world — US East, US West, Europe, Asia Pacific, and others. When an issue happens in one region, it affects only the resources you have in that region. If you have built redundancy into your process by running it in multiple regions, an issue in one region may not take down your whole service.

Some issues are account-wide, meaning they affect your ability to use AWS services no matter which region you are in. This might happen if there is a problem with your account itself, or if there is a widespread AWS problem that spans multiple regions. These are rarer but more serious.

If you are troubleshooting an issue, always check which region you are looking at. You might be checking the status of US East when your resources are actually in Europe. The AWS console shows you which region you are currently viewing in the top right corner.

What happens to your bill during an AWS issue

If AWS has an issue that makes a service unavailable, you are usually not charged for that time. AWS publishes a Service Level Agreement (SLA) for each service that spells out what happens if the service does not meet its uptime promise. If an outage is caused by AWS, you may be owed a service credit — a refund or credit toward your next bill.

However, you have to request the credit. AWS does not automatically explore it. You request it through the AWS Support console, and you will need to provide details about how the outage affected you. The credit amount depends on how long the outage lasted and how much uptime the SLA promised.

If the issue is on your side — your misconfiguration, your code, your quota — you are still charged for any resources you are using, even if they are not working. This is why it is important to diagnose the problem quickly and fix it.

Common AWS issues and what causes them

Some issues come up again and again. Running out of EC2 instances (the virtual servers AWS rents you) is common when you are scaling up an process. You hit your account quota and cannot launch more instances until you request an increase. Security group misconfiguration is another frequent one — you set up a rule wrong, and traffic cannot reach your server, so it looks like the server is down when it is actually just unreachable.

Database connection issues happen when your RDS database is not accepting connections. This might be because the database is out of storage, or because you have hit the maximum number of connections allowed, or because the security group is blocking traffic. IAM permission issues are also common — your code has the right to do something in theory, but the IAM role attached to it does not actually grant that permission.

Occasionally, AWS itself has a real outage. The most famous recent example was a problem with AWS's US East 1 region in December 2021 that lasted several hours and affected many major websites. But these are relatively rare. Most of the time, when something is not working, the problem is in your own setup.

What to do when you find an AWS issue

Start by checking the AWS Health Dashboard and the AWS Status Page. If they show a problem, there is not much you can do except wait for AWS to fix it. You can check back periodically to see if the issue has been resolved. If you have a support plan with AWS, you can open a support case, though AWS support cannot fix the underlying infrastructure problem any faster than their engineers are already working on it.

If the dashboard is green, assume the problem is on your side and start diagnosing. Check your security groups, your IAM permissions, your service quotas, and your process logs. Look at the CloudWatch metrics for your resources to see if anything unusual is happening. If you still cannot find the problem, you can open a support case with AWS, and an engineer can help you dig deeper.

Document what you find. Write down when the issue started, what service was affected, what you were trying to do, and what error message you saw. This information will help you if you need to open a support case, and it will also help you recognize the issue if it happens again.

Frequently Asked Questions

How do I know if an AWS issue is affecting me right now?

Log into your AWS console and go to the AWS Health Dashboard. It shows real-time status for all AWS services and regions. If you see a yellow or red indicator for the service you are using, AWS is having a problem. You can also check status.aws.amazon.com without logging in to see the public status page.

Will I be charged if AWS has an outage?

You are not charged for the time the service is unavailable, but you have to request a credit. Open a case in the AWS Support console and provide details about the outage and how it affected you. AWS will review your request and issue a credit if the outage was on their side and met the terms of the SLA.

What should I do if the status page is green but my service is still down?

The problem is likely on your side. Check your security groups, IAM permissions, and service quotas first. Look at your CloudWatch logs and metrics to see if your process is crashing or if traffic is being blocked. If you still cannot find it, open a support case with AWS and provide the details you have gathered.

Can an issue in one AWS region affect my service in another region?

Not usually. Issues are typically regional, so if you have resources in multiple regions, an outage in one region should not affect the others. However, if your process depends on a shared service (like a database in one region that your servers in other regions connect to), a regional issue could still affect your whole process.

How long do AWS issues usually last?

It varies widely. Minor issues might be resolved in minutes. Larger outages can last hours. The AWS Status Page shows a history of past issues and how long each one lasted, which can give you a sense of what to expect. Most issues are resolved within a few hours.